Reconstructed historical example · frozen pregame inputs

Built only from inputs that existed before kickoff. This archived example does not update.

Inputs frozen as of Tue, Sep 8, 1:55 AM EDT · kickoff Sun, Sep 13, 1:00 PM EDT

Sample game ▸ 2026 Week 1

Baltimore Ravens at Indianapolis Colts

A reconstructed pregame example, built from historical market data and governed model inputs that existed before kickoff, shown the way the member Desk presents these views. It is not a record of what members saw at that moment.

Market state

Consensus across the books tracked, observed Mon, Sep 7, 4:06 AM EDT.

Full-game total

Consensus total: 48.0
Books: 11
Movement: −0.5 from 48.5
Nearest key total: 47 · 1 point away

Movement is measured from the first observation, Sun, Jul 12, 5:26 AM EDT.

Full-game moneyline

Indianapolis Colts (home): +150
Baltimore Ravens (away): −180
Books: 11

Consensus prices are market context — what the books were pricing. They are not an FDI view.

Governed model context

Each view states when it was generated. Not every game has every view, and nothing is estimated in the place of a missing one.

Projected game total

46.2

Generated Sat, Sep 5, 6:45 PM EDT

Model limitation: weather data was not available to the model.

FDI's model view is not an FDI Position. It describes what the model expects for the game total, not what to do about it.

Moneyline win probability

Indianapolis Colts: 58.6%
Baltimore Ravens: 41.4%

Generated Tue, Sep 8, 1:55 AM EDT

FDI's moneyline model view is not an FDI Position. It reports a probability, not what to do about it.

Intel Score

66 / 100

The Intel Score is a 0–100 measure of how strongly the available evidence supports the way a read was built.

An Intel Score describes how a read was built, not whether FDI holds a position on the game. It is not a win, cover, or hit probability, and not an expected return.

Moneyline Lean

Lean · Indianapolis Colts (home)

Directional model context, not an FDI Position or an instruction.

Model probability: 58.6%
Captured market (no-vig): 38.2%
Model-market difference: +20.4 pp
Approved books: 5

Market observed Mon, Sep 7, 4:06 AM EDT

In this frozen reconstruction, "current" means current at the Lean evaluation instant, with the market observed at the time shown above.

A Lean is a mechanical comparison, not a forecast of the result. FDI takes the win probability its frozen moneyline model produced before the season, compares it with the latest captured approved-book market, and reports a Lean only when at least 3 approved two-sided books are present, the difference is at least 5.0 percentage points, and the best current approved price still sits below the model's own probability. The rule was fixed in advance for the 2026 season and is not tuned to past results.

Moneyline Lean is separate from FDI Position.

How to read this

Market figures are the books' consensus at the capture time shown. They are not an FDI view.
The projected total and the win probability are what FDI's governed models expected before kickoff. They describe an expectation, not an action.
The Intel Score is a 0–100 measure of how strongly the available evidence supports the way a read was built. It is not a win, cover, or hit probability, and not an expected return.
A Moneyline Lean is directional model context, not an FDI Position or an instruction. Many games have no Lean.
This is one archived reconstruction from pregame inputs. It does not update, and it does not show how the game ended.

Members see governed views like these on the Desk for games where eligible reads exist.

Create your account

Player Prop Intelligence · demonstration

Fictional demonstration. Harbor City and Ridgeline are invented teams, the players are invented, and every value below is frozen demonstration data. It is not a real game, not an FDI projection, and it does not update. It shows how the member Player Props view is laid out.

Members see one board per game, organized by player. For each supported family — passing yards, receptions — a row shows the FDI projection, frozen two hours before kickoff, beside the same player's current sportsbook line, and the difference between them. The difference is simple arithmetic, not a read.

Quarterbacks

Quarterbacks: demonstration projection, market line and difference
Player	Passing yards
Owen CalderHBC · QB	FDI projection 241.6Market 236.5Difference, projection minus line, +5.1
Luis FerranteRDG · QB	FDI projection 219.8Market 224.5Difference, projection minus line, -4.7

Receivers

Receivers: demonstration projection, market line and difference
Player	Receptions
Theo MarshHBC · WR	FDI projection 5.8Market 5.5
Nate VossHBC · TE	FDI projection 4.1Market 3.5
Andre KellRDG · WR	FDI projection 6.3Market 5.5
Cal BrennanRDG · WR	FDI projection 3.2No current line

Running backs

Running backs: demonstration projection, market line and difference
Player	Receptions
Miles OrtegaRDG · RB	FDI projection 2.9Market 2.5

When a line is not current

The same quarterback, with a market observed the day before and no longer current. The line is still shown, with its time and a “not current” label, but no difference is calculated against it.

241.6Market 236.5 · not current

A player dossier

Selecting a player opens their dossier: every supported family for that player, with the line's prices, how many books quote it and when it was observed. Where a model estimates the chance of finishing above the exact number the current line is set at, the dossier shows that probability too. It is model context, not a recommendation.

Owen Calder · Harbor City · QB1

Passing yards

FDI projection: 241.6
Market line: 236.5
Over / Under: -110 / -110
Observed: 12:41 PM ET
Difference: +5.1

Theo Marsh · HBC · WR

Receptions

FDI projection: 5.8
Market line: 5.5
Over / Under: -115 / -105
Observed: 12:41 PM ET
Difference: Not shown
Model probability: Shown only at a modeled threshold

Membership

Members see governed views like these on the Desk for games where eligible reads exist, alongside the market's own state.

Create your account Membership plans and pricing